Implementation
Langdon: OK, we're going to report.
Your assignment is to work with the ideal system model, identify at least three ways to improve it, and also to design a process to implement it.
Please work with your teammates and discuss the following questions:
Step 1: About 45 minutes
Assume that your team is in charge of implementing the model as a single system of care throughout California, between today and 2015.
How would you go about doing that?
Please identify at least ten major milestones that would be necessary for implementation, and show them on your white board in sequence on a timeline.
For each milestone, identify who would be the key individuals or organizations that would play the key role in that activity, and what their responsibility would be.
Please let us know when you’re finished with Step 1, and we’ll bring the instructions for the additional steps.
Step 2: About 45 minutes
Sometimes when you want to make significant changes to a large and complex system, you can design smaller scale experiments to test your ideas before you commit to a full scale change.
Identify 5 – 10 such experiments that could be done in this system that would help us know the best ways to proceed. Prioritize them in order of importance.
Step 3: About 20 minutes
The Lucile Packard Foundation for Children’s Health would like your input into what role it should play in implementing the ideal system. If you had $2 million to invest in this effort in each of the next 5 years, how would you invest it?
You will share your work with the other groups when you’re finished.
Team 1 Report Out
Our definition of the medical home is comprehensive and it includes community services. We wanted to identify the pivotal steps - those that might have more impact. One that came out quickly was to change the way that primary care providers take responsibility for care coordination. This system will only work if PCPs are more capable and systems can facilitate high quality care in community settings.
Community services need to be linked to PCPs and be more available.
The other component we thought was pivotal was to build a way of generating an enhanced public presence - bringing stakeholders together and to advance the needs of children with special needs in a public setting.
The mechanisms we identified to implement these processes fell in a time line that gets the stakeholders together - county, state, specialty care, family advocacy and care givers - and then create a pilot implementation (not demonstration - but pilots that are focused on those key pivotal spots).
Focused on enhancing the capabilities of PCP docs and non-docs to deliver care and to look at the funding mechanisms.
The model we were handed has a single payer - the structural issues and the political implications are going to be significant. We need to be purposeful about the design of these things that inform the conversation of a single payer. It should be approached more comprehensively and build some support for this.
When a kid in Santa Clara county is on Medicaid - they are covered; but they could also be covered by CCS. There are county based health programs that look like private health plans.
One pilot might be to do a flow of payment model. The hospital would get so much money and they could pay for the care for 6 months or a year. By coming up with a unified payment mechanism rather than just blending we could pay the hospital a larger sum at the beginning. If that was the pilot that would answer the essential questions - if the real issue is how to get primary care providers to facilitate their role in the care delivery that could be another thing to pilot.
What this pilot should do is not to mimic what's been done in the literature. My fear is that the Foundation would succeed in doing things that have already been done before.
We need to focus on these strategic problems.
We talked about a new regional entity to be held accountable to the quality of care and the whole system for these kids.
To make this enforceable in any real way it would have to have a component that speaks to the financial components.
We felt the role of the foundation should be very focused. $2 million a year is an inadequate amount to do the implementation of just about anything real. The foundation could convene and lubricate the system with strategic funding. Use it's resources to create an integrated and unified approach to these pilot programs. Generate the list of critical problems and do the evaluation but NOT expending money on direct service provision and altering the delivery mechanisms.
The issue is scale.
Your assignment is to work with the ideal system model, identify at least three ways to improve it, and also to design a process to implement it.
Please work with your teammates and discuss the following questions:
Step 1: About 45 minutes
Assume that your team is in charge of implementing the model as a single system of care throughout California, between today and 2015.
How would you go about doing that?
Please identify at least ten major milestones that would be necessary for implementation, and show them on your white board in sequence on a timeline.
For each milestone, identify who would be the key individuals or organizations that would play the key role in that activity, and what their responsibility would be.
Please let us know when you’re finished with Step 1, and we’ll bring the instructions for the additional steps.
Step 2: About 45 minutes
Sometimes when you want to make significant changes to a large and complex system, you can design smaller scale experiments to test your ideas before you commit to a full scale change.
Identify 5 – 10 such experiments that could be done in this system that would help us know the best ways to proceed. Prioritize them in order of importance.
Step 3: About 20 minutes
The Lucile Packard Foundation for Children’s Health would like your input into what role it should play in implementing the ideal system. If you had $2 million to invest in this effort in each of the next 5 years, how would you invest it?
You will share your work with the other groups when you’re finished.
Team 2 Report Out
OK. We were the system model implementation team. Our job was to come up with some steps to say how it was done. This is our timeline and our steps.
This starts today with advocacy, lobbying, and a preparation effort - we need to put together the data and the knowledge to influence the people that need to be influenced (the state legislature).
Then put together a champion group and a proposal that would lead to a law or a task force at the state level. They would be charged with designing the system - including a vast array of participants. They would develop the principles and the quality standards to implement a pilot.
The step in between would be to create a new administrative body at a governmental level - because one of our pieces is a new funding approach and that would require a new body to oversee that.
The pilot would be at a level of structural, budget, process, recommendations, etc. to all people that have a role of implementing care in this state.
The pilot isn't defined any further. There is some suggestion that this is a comprehensive medical home model. The pilot should also address some structural things at a governmental level.
What's the scale of the pilot? Some people suggest to do it at a county level and others say a county is too small. Some people think it could be a hospital in a rural area. We didn't answer the question of scale. We punted on that.
Qualifying conditions is impossible to do. That creates a quagmire and is un-reconcilable. The solution is to include everyone - all children. There was a pretty good case made for this.
What does it mean to include all children when 15% have urgent and intense needs? What does it mean to include kids that don't look like the ones we have now?
The argument was that it's easier to create a whole system that is inclusive and then monitor the individual groups within relative to some quality standards.
Here's some experiments:
- Include a county by county study to understand the best practices and things that are going on at a county by county level.
- Create a local mini task force at some level to test the ability to create a state wide task force down the road.
- There was a suggestion to do some message testing - market research to figure out what messages would make a difference.
- Run a care coordination trial to extract data to make the case - not to perfect care coordination. How would it be different from other care coordination experiments? Don't know.
- Focus a trial on a condition - and that one too we weren't too excited about.
- Look at other advocacy efforts and try to learn from those and find the keys to success there.
The conversation turned to efficiency and cost savings. One of the things that could be brought to bear might be to model costs and learn the inefficiencies.
Can we show the consequences on families in California? How much data can we put together to give a complete story? What are the economic consequences of this population on the state? What are the costs of having many different programs addressing this?
Creating the Al Gore's Inconvenient Truth presentation to raise awareness was something that could be done with the Foundation money in the short term.
Then we could create a state wide task force two years from now.
In our group we settled on the pilot concept to test different approaches. Testing a medical home pilot and compare it to another model?
We started down the path of listing everyone that needed to be involved and then we stopped. We got very focused in on the first little bit and trying to influence the thoughts of people in the state level. We don't know what the appetite is at a state level.
In your break out team, please work with the ideal system model to design the strategy for implementing the system. Please follow these steps in your design process:
Step 1: About 15 minutes
State the goal of the system, and identify how performance of the system would be measured in qualitative and quantitative terms.
Step 2: About 45 minutes
- Identify three completely different strategies or approaches to implementing the system. (Yes, three) Consider, for example, top-down strategies vs. bottom-up. Consider also political, financial, and public relations approaches.
- Identify the five key steps that would constitute each strategy.
Please let us know when you’re finished with Steps 1 & 2, and we’ll bring the instructions for the additional steps.
Step 3: About 30 minutes
- At this point you should have identified 3 strategies and 5 key steps for each one. Now select the one strategy that seems the most likely to succeed and expand on the 5 implementation steps to show in more detail exactly how it would be implemented.
Also, identify the key organizations that would have to participate in its implementation, and specify what they would do by identifying the role (s) of each.
- Determine the steps that would be needed in order to shift from the current system to the ideal system.
What would you do first? Second? Etc.
- Which stakeholders would have to be convinced that the strategy was a good one? How would you convince them?
Step 4: About 20 minutes
The Lucile Packard Foundation for Children’s Health would like your input into what role it should play in implementing the ideal system. If you had $2 million to invest in this effort in each of the next 5 years, how would you invest it?
You will share your work with the other groups when you’re finished.
Team 3 Report Out
We had three strategies. Strategy number one is small scale, bubble up. Piloting an idealized system in a constrained geographic area which includes all payors in that county (this would really be employers).
Strategy two was to leverage the 1115 wavier process to describe and provide a vehicle for system change.
Strategy three is similar to the Comprehensive Perinatal Services Program: creating a system based on care coordination with fee-for-service payment mechanisms for participating providers. Care coordination could be paid for by fees from plans, providers and possibly funded grants or matching funds.
We decided to drill down on strategy two with some attention being paid to strategy three.
We would like to influence the creation of a special needs plan, potentially available to all children in the state of California who qualify. It covers the whole child. Because of the 1115 wavier, it covers all Medicaid and CHIP. Commercial plans can buy in or not. It’s a form of reinsurance.
There was a discussion about the common health plan in MA which allows any family to buy in on an individual basis. We might be able to use aspects of this approach.
We hung a few other ideas on this strategy. There would be no change to the delivery system. There would be a new ASO (Administrative Services Org) that would replace EDS. There was the idea that the provider in this plan, besides meeting the criteria, would also participate in a Medical Home Chronic Care Management Program.
That’s the overall approach.
There were some areas of discussion. What is the role of the county and its health systems in this plan? There is a county infrastructure that’s involved in the delivery and what happens to that? Each county has different infrastructures. A few have managed care plans. Most don’t. Is this a one size fits all plan?
Eligibility would have to be equal to or possibly more focused than the current CCS.
We also discussed triggers for some form of cost managed care.
There’s a lot of work to do for this to succeed such as engaging with families, CCS provider networks and legislators. There needs to be a planning process for 1115 wavier and this requires an unbiased convener.
This becomes a very political process, and that may make it risky for the foundation to take on.
In your break out team, please work with the ideal system model to design the strategy for implementing the system. Please follow these steps in your design process:
Step 1: About 15 minutes
State the goal of the system, and identify how performance of the system would be measured in qualitative and quantitative terms.
Step 2: About 45 minutes
- Identify three completely different strategies or approaches to implementing the system. (Yes, three) Consider, for example, top-down strategies vs. bottom-up. Consider also political, financial, and public relations approaches.
- Identify the five key steps that would constitute each strategy.
Please let us know when you’re finished with Steps 1 & 2, and we’ll bring the instructions for the additional steps.
Step 3: About 30 minutes
- At this point you should have identified 3 strategies and 5 key steps for each one. Now select the one strategy that seems the most likely to succeed and expand on the 5 implementation steps to show in more detail exactly how it would be implemented.
Also, identify the key organizations that would have to participate in its implementation, and specify what they would do by identifying the role (s) of each.
- Determine the steps that would be needed in order to shift from the current system to the ideal system.
What would you do first? Second? Etc.
- Which stakeholders would have to be convinced that the strategy was a good one? How would you convince them?
Step 4: About 20 minutes
The Lucile Packard Foundation for Children’s Health would like your input into what role it should play in implementing the ideal system. If you had $2 million to invest in this effort in each of the next 5 years, how would you invest it?
You will share your work with the other groups when you’re finished.
Team 4 Report Out
Our assignment was to come up with three strategies for the ideal system and then steps for each one.
Our spectrum of strategies ran from whether we should do something statewide or do something that was a pilot, or something in between. We identified steps for the statewide strategy and the pilot.
We started with a federal wavier in order to do a statewide pilot. The other steps are pretty intuitive:
- Defining eligibility criteria
- Developing data management system
- Establishing financing contributions
- Defining outcome measures
- Using system efficiency savings to improve reimbursement and bring some of the providers back into the system
Then we focused our efforts on the local pilot strategy in a county. We identified five primary action steps. This was the model that we chose to jump into more deeply. We decided on a pilot strategy at a county level.
One of the primary steps that one would have to consider in doing a pilot were: are you choosing pilots to optimize success or do you want a true test in a difficult environment? For example, LA county would be a true test because of its inherent complexity, but San Diego might have better chances for success because there is one of everything.
If you do a pilot, you need to consider which county you’re going to work in. Rural or urban? Accessibility of a children’s hospital is also important. Some counties have varying assets and deficiencies as well.
After we select a target county, we need to identify and convene the stakeholders. Some were county-specific but others like Regional Centers were from a broader geography. We identified state, federal, county and regional stakeholders as well as payors, families and physicians.
Next we looked into a model of care. We wanted to have the Medical Home in our program, however it’s defined, and we wanted an expanded role for care coordination. Medical Home will likely vary by child and condition. We didn’t want to be prescriptive. Care coordination is currently provided by lots of groups but they tend to be gatekeepers for their own organizations. Regional Centers are examples. When we talk about care coordination we mean a broader role.
We talked about the need for bringing legislation in place to allow for some of these pilots to occur.
We wanted to make sure that there was front end support for Care Coordination costs in order to promote efficiencies in other parts of the system. That’s one thing we thought the Lucile Packard Foundation for Children's Health could provide support for. This could happen in a few ways. One would be small grants to providers to serve as care coordinators. Another would be a care coordination utility. We also considered the funding of facilitating the process of improving efficiencies in the provision of care.
We also talked about the possibility of supporting the investment in data collection and mining and supporting family advocacy. These are also candidates for Lucile Packard Foundation for Children's Health investment.
Comments
The same central ideas have emerged in every group we have sat in so far. Some groups have a greater focus on the legislative aspects of the challenge.
We were thinking of using the convening more for buy-in and to help with what outcomes we should be looking for.
Everyone is kind of on board with the pilot approach. How would you resolve the issue of choosing a pilot? Another team settled on an early step of convening a county task force. It might not be a single pilot in a single county. We were not piloting a change in the system, but a process for coming to consensus about what changes should be made in the system.
Nobody has discussed the fact that we’re in the middle of national health care reform and statewide health care reform. It seems to me that somehow our effort and those efforts ought to be integrated and coordinated. Health care may be a moving target in some ways.
Langdon
Please come in and take a seat. The purpose of getting back to together in a large group to see how you feel the alignment is between the four groups. Is there alignment or is there divergence?
I felt like there was convergence on the two sides of room but divergence between the two sides of the room.
The timeline was different on each side - we were looking at short timelines on one side and the other side was a much longer time frame.
There were elements on each side that went very nicely with the other. The waiver idea was one.
A key question is pilot or not to pilot. There were teams that advocated each way. Not every group explicitly addressed the issue of all-in (all kids are in). Another group focused on more focused on the CCS group of kids.
Thanks for that little bit of feedback. Please go back into your groups and do this convergent assignment.